8f05a7c188
- Update ASR, face, OCR, pose processors - Add release pre-flight check script - Add synonym generation, chunk processing scripts - Add face recognition, stamp search utilities
88 lines
2.4 KiB
Python
Executable File
88 lines
2.4 KiB
Python
Executable File
#!/opt/homebrew/bin/python3.11
|
|
"""
|
|
pyannote.audio 測試腳本
|
|
測試說話人分離功能
|
|
"""
|
|
|
|
import sys
|
|
import json
|
|
import os
|
|
|
|
def test_pyannote(audio_path, output_path):
|
|
"""測試 pyannote.audio 說話人分離"""
|
|
|
|
print(f"[pyannote] Testing on: {audio_path}")
|
|
|
|
try:
|
|
from pyannote.audio import Pipeline
|
|
|
|
# 載入模型
|
|
print("[pyannote] Loading model...")
|
|
pipeline = Pipeline.from_pretrained(
|
|
"pyannote/speaker-diarization-3.1"
|
|
)
|
|
|
|
print("[pyannote] Model loaded successfully")
|
|
|
|
# 執行說話人分離
|
|
print("[pyannote] Performing speaker diarization...")
|
|
diarization = pipeline(audio_path)
|
|
|
|
# 收集結果
|
|
segments = []
|
|
for turn, _, speaker in diarization.itertracks(yield_label=True):
|
|
segments.append({
|
|
"start": turn.start,
|
|
"end": turn.end,
|
|
"speaker": speaker
|
|
})
|
|
|
|
# 輸出結果
|
|
result = {
|
|
"audio": audio_path,
|
|
"num_speakers": len(set(s["speaker"] for s in segments)),
|
|
"segments": segments
|
|
}
|
|
|
|
with open(output_path, "w") as f:
|
|
json.dump(result, f, indent=2)
|
|
|
|
print(f"\n=== 測試結果 ===")
|
|
print(f"說話人數量：{result['num_speakers']}")
|
|
print(f"片段數量：{len(segments)}")
|
|
print(f"輸出檔案：{output_path}")
|
|
|
|
# 顯示前 5 段
|
|
print(f"\n前 5 段：")
|
|
for i, seg in enumerate(segments[:5], 1):
|
|
print(f" {i}. [{seg['start']:.2f}s - {seg['end']:.2f}s] {seg['speaker']}")
|
|
|
|
return True
|
|
|
|
except Exception as e:
|
|
print(f"[pyannote] Error: {e}")
|
|
import traceback
|
|
traceback.print_exc()
|
|
|
|
# 寫出錯誤資訊
|
|
result = {
|
|
"audio": audio_path,
|
|
"error": str(e)
|
|
}
|
|
with open(output_path, "w") as f:
|
|
json.dump(result, f, indent=2)
|
|
|
|
return False
|
|
|
|
|
|
if __name__ == "__main__":
|
|
if len(sys.argv) < 3:
|
|
print("Usage: python3 test_pyannote_audio.py <audio_path> <output_path>")
|
|
sys.exit(1)
|
|
|
|
audio_path = sys.argv[1]
|
|
output_path = sys.argv[2]
|
|
|
|
success = test_pyannote(audio_path, output_path)
|
|
sys.exit(0 if success else 1)
|